Because of a sharp increase in deaths among older adults with dementia during the onset of the COVID-19 pandemic, the rate of older adults with the disorder declined faster from 2011 to 2021 than it did through 2019, according to a University of Michigan study.

With heightened social isolation brought about by the onset of the pandemic, many speculated that dementia would become more commonplace among older adults, dampening the pre-pandemic trend toward lower rates. The researchers, led by University of Michigan Institute for Social Research scientist Vicki Freedman, found some support for this idea in 2020, but by 2021, the pattern was better described as an accelerating downward trend.

“Before the pandemic, dementia rates were declining in the U.S. At the onset of the COVID-19 pandemic, the rate of dementia increased slightly, as many expected,” Freedman said. “But these increases were swamped by sharp increases in deaths among older adults living with dementia that occurred in 2020-2021. As a result, the downward trend in dementia for 2011 to 2021 was even steeper than the pre-pandemic trend.”

The rate of dementia declined from 11.9% in 2011 to 9.2% in 2019, and, after a small uptick to 9.6% in 2020, dropped to 8.2% in 2021. In average yearly terms, the decline was 2.8% from 2011 through 2019, and 3.1% from 2011 through 2021.

The study results are published in the Journals of Gerontology: Medical Sciences.

Freedman and co-author Jennifer Cornman of Jennifer C. Cornman Consulting used the National Health and Aging Trends Study to examine over time the percentage of the older population with dementia, developing dementia, and dying with and without dementia.

The researchers used information from approximately 48,000 annual interviews with adults 72 and older that took place from 2011 to 2021. Funded by the National Institute on Aging, part of the National Institutes of Health, NHATS is designed to study trends in disability and functioning of the older U.S. population, including cognitive functioning.

“Research that assesses national trends in dementia across important life events such as the COVID-19 pandemic requires robust data like this,” said John Phillips, acting deputy director of NIA’s Division of Behavioral and Social Research. “NHATS captures social, behavioral, economic and environmental data longitudinally in a nationally representative population sample to track trends in older adults so that researchers can understand changes over time and in turn better inform policies and practices to prepare for the future.”

Each year, NHATS participants were identified as having dementia if they had a reported diagnosis, screened for dementia based on answers from a proxy reporter, or scored low on multiple tests evaluating cognitive function.

The research also explored reasons for the decline. Previous work by Freedman and colleagues through 2015 pointed to the larger, better-educated generation entering their 70s in the mid-2010s as driving a good portion of the decline. Freedman found that now these factors are having less of an effect.

“We see that age and education are accounting for a smaller fraction of the decline in dementia trends once we incorporate the post-COVID period,” Freedman said. “Because of these longer-term population shifts, the U.S. can no longer expect sustained declines related to the ‘younging’ of the population or increases in education.”

Freedman says shifts in the composition of race and ethnicity of the U.S. population could also slow declines in dementia prevalence in the future. Rates of dementia are often higher among minority race and ethnic groups, which will make up a larger share of the older U.S. population in the future.

Freedman notes there are lingering questions about the impact of the COVID-19 pandemic on population-level dementia trends.

“The longer-term effects are still unclear, including the consequences of having an increasing share of the older population experiencing multiple bouts of COVID-19 infection or lingering health effects,” Freedman said. “These are important questions for future research to address.”

This work was supported by the National Institute on Aging, part of the National Institutes of Health, grant numbers U01AG032947 and P30AG012846.